Covering Disruptive Technology Powering Business in The Digital Age

Written by: Martin Dale Bolima, Tech Journalist, AOPG.

A small claims adjudicator in British Columbia—the Civil Resolution Tribunal (CRT)—has determined that an Air Canada chatbot had given a grieving customer misleading information about the airline’s bereavement policy and has thus ordered it to pay the complainant, Jake Moffatt, USD $812.

The USD $812 is meant to cover the difference between Air Canada’s bereavement rate and the USD $1,630.36 Moffatt paid to buy full-price tickets to and from Toronto when the family visited the wake of their just-departed grandmother.

According to reports, Moffatt’s grandmother died on Remembrance Day 2022. The young man visited the airline’s website that same day to check which documents he had to submit to qualify for bereavement rates and whether or not the refund could be granted retroactively.

In his search, Moffatt got to interact with the Air Canada chatbot, who advised him: “If you need to travel immediately or have already travelled and would like to submit your ticket for a reduced bereavement rate, kindly do so within 90 days of the date your ticket was issued by completing our Ticket Refund Application form.”

This was all Moffatt had to see to convince him to buy tickets at their full price. After all, retroactive refunds were allowed, at least according to the Air Canada chatbot.

A Glaring Disconnect—But Who’s at Fault?

Evidently, the airline had no such policies on completed travel. Meaning, the refund could not be given retroactively. This was explained on the website itself but on a different page—likely the one the Air Canada chatbot did not know about or ingest just yet. It was explained to Moffatt as well when he tried to recoup the refund, with the airline admitting to the wrong information provided by its digital agent and promising to update the bot.

Moffatt sued as a result. In its defence, Air Canada argued—perhaps quizzically—that its chatbot was a “separate legal entity” and was therefore solely responsible for its actions.

Adjudicator and CRT member Christopher Rivers, who ruled in favour of Moffatt, did not buy Air Canada’s defence, countering that the chatbot, being part of the Air Canada website, is also the airline’s responsibility.

“This is a remarkable submission,” Rivers wrote in his decision, which was released in the second week of February. “While a chatbot has an interactive component, it is still just a part of Air Canada’s website. It should be obvious to Air Canada that it is responsible for all the information on its website. It makes no difference whether the information comes from a static page or a chatbot.”

Moreover, Rivers argued it was Air Canada’s responsibility to ensure its chatbot provided accurate information.

“I find Air Canada did not take reasonable care to ensure its chatbot was accurate,” Rivers added. “While Air Canada argues Mr Moffatt could find the correct information on another part of its website, it does not explain why the webpage titled ‘Bereavement travel’ was inherently more trustworthy than its chatbot. It also does not explain why customers should have to double-check information found in one part of its website on another part of its website.”

Chatbots: Boon or Bane?

The Air Canada chatbot is yet another proof of the growing popularity of digital agents in commerce, as “businesses are beginning to leverage new channels to respond to their customers such as live chat, messenger apps, and social networks,” explained Anand Venkatraman, COO at CleverTap, to Disruptive Tech News in an exclusive interview last year. With this connectivity, Venkatraman added, “more organisations are seeing the benefits of chatbots to address and respond to the needs of their customer.”

This need to leverage new channels is probably one reason Air Canada introduced a chatbot to augment its website. And, to be fair, the Air Canada chatbot did respond to Moffatt’s need at that time—albeit with the wrong information. This incident underscores the other side of the fence when it comes to using digital agents, who can be a bane to companies when they provide inaccurate responses (as was the case with Moffatt), fail to provide exactly what customers are looking for or drag interactions that lead to nowhere.

Air Canada’s defence of its chatbot being its own entity and apart from the airline also raises important questions:

1. Are chatbots really separate from the companies they are representing?
2. How much oversight are companies expected to take on their chatbots?
3. To what extent are companies responsible for the actions of their chatbots?

Based on the CRT ruling on Air Canada’s chatbot snafu, it appears that no, chatbots are not separate entities from the companies they are representing. These same companies, evidently, are also duty-bound to exercise oversight on their digital agents and make sure they are operating as intended—and providing only relevant, timely, and most important of all, accurate information.

But do these hold true everywhere else? What if something similar happens in, say, Malaysia or Singapore? What if a chatbot messes up in the Philippines or Indonesia?

The Air Canada Chatbot Incident Is A Lesson for All 

The case of the Air Canada chatbot—along with the questions it raises—is one that company higher-ups need to ponder as reliance on chatbots continues to grow in this digital age.

In particular, companies will need to figure out foolproof mechanisms to ensure that they won’t repeat the mistake of the Air Canada chatbot because such errors can have significant ramifications. In the case of Canada’s flag carrier, the financial damage is almost negligible—especially given the company’s USD $21.833 billion revenue for 2023.

The takeaway here is simple: Chatbots are fallible; they can make mistakes. And it might just be a matter of time before one makes an error that will cost a company millions.

It will be up to you and your team, then, to make sure it is not yours.

(0)(0)